ever, is manifestly a fiction and has been generally rejected by recent scholars. The following considerations prove that it is false: (1) As will appear from our analyses of the romances and from the discussion below, there is no likelihood that even the Lancelot is the work of one man, much less three romances so different in power, style, and tone as the Lancelot, Queste, and Mort Artu. (2) No one has ever claimed for the Lancelot an earlier date than the last decade of the twelfth century generally it is dated later but even in 1190 Map was a man of about fifty years of age and it is in the highest degree unlikely especially under mediaeval conditions that he should have taken to the writing of romances at that time of life. (3) Map himself expressly

celot, but one cannot say how often, until further collations of that branch have been made. For MSS. of the Vulgate cycle cp. H. O. Sommer, Vulgate Version of the Arthurian Romances, I, pp. XXIII - XXXII. For the Lancelot MSS. cp. besides, the Vorwort to G. Bräuner's Der altfranzösische Prosaroman von Lancelot del Lac Branche. Marburger Beiträge zur romanischen Philologie, Heft II, and for those of the Mort Artu, Bruce's edition of that romance, pp. XII, ff. In the last-named list the following MSS. are omitted: 122, f. fr. (B.N.) which, like MS. 342, shows an unabridged text for the passage, pp. 102 ff. (Bruce's ed.) and the Vatican MS. Palat. 1967. For the latter (recently discovered), which contains only the Mort Artu, cp. Romania, XLVI, 151 (1920). F. Lot has in preparation a special work on the MSS. of the Vulgate cycle.

3

Cp., for example, Birch-Hirschfeld, Sage vom Gral, pp. 227ff. (Leipzig, 1877), G. Paris, Manuel, 62, and Sommer, Zs. f. rom. Ph., XXXII, 336 (1908). As will be seen, however, below, p. 371, note 8, the old idea of Map's connection with the cycle has been revived in a modified form by certain scholars.

4

For the life of Map, see H. L. D. Ward's Catalogue of the Romances in the Department of Manuscripts in the British Museum, I, 734ff. (1883), and the article on him in the Dictionary of National Biography. These are better than the fuller, but antiquated, life by Georg Phillips, Walter Map: Ein Beitrag zur Geschichte König Heinrichs von England und des Lebens an seinem Hofe, Sitzungsberichte of the Vienna Academy of Sciences, Philos. hist. Klasse, X, 319ff. (1853). Concerning earlier views about W. Map and the Lancelot, cp. Graesse, Lehrbuch einer allgemeinen Literärgeschichte, Band II, Abtheilung III, pp. 188 ff. Dresden and Leipzig (1842). Hesperia, Ergänzungsreihe: 8.

24

states in the oft-quoted words preserved by Giraldus Cambrensis that he wrote little or nothing. According to Giraldus, Map said to him: "Multa, magister Giralde, scripsisti, et multum adhuc scribitis: et nos multa diximus. Vos scripta dedistis et nos verba." He goes on to say that, although the writings of Giraldus were of much more importance than his (Map's) words, (i. e. spoken words), yet, because his words were in the vernacular, which everybody understood, they had brought him greater profit than all of his friend's Latin."

The implication of this statement of Map's is borne out by the little that we know of his life and activities, his only authentic literary production being the De Nugis Curialium (Courtiers' Triflings), which is a collection of tales and anecdotes loosely thrown together. In his own time his reputation was that of a wit and satirist, and there is no evidence that he possessed even Hiberniae Expugnatio, Rolls Series edition of Giraldus's Works, V, 410f. (1867).

6

P. Paris, Romans de la Table Ronde, I, 472, tried to get around the plain meaning of these words by interpreting scribere as writing in Latin, dicere as writing in the vernacular, but there is no such distinction as this elsewhere. Cp. Birch-Hirschfeld, p. 229. Brugger, too, op. cit., p. 93, note 49, denies any importance to these words reported by Giraldus; for, as a matter of fact, he says, Map did write the De Nugis, and he may have counted the (supposed) Lancelot romance as just as little worth mentioning as that work. The two works, however, would certainly be on a different plane, for the De Nugis, as said above, is a mere disconnected collection of anecdotes and tales and there is no likelihood that Map ever attempted to make it generally known, seeing that it exists in only one fifteenth century) MS. Cp. the recent and best edition of the book by M. R. James (Oxford, 1914). Indeed, James Hinton, whose study of the De Nugis, viz., "Walter Map's De Nugis Curialium, its plan and composition," PMLA, XXXII, 81ff. (1917), is the ablest we have, concludes that Map left this book in the condition of loose fragments, which were put together by some one else after his death. So, too, Henry Bradley, English Hist. Review, XXXII, 400 (1917).

It should be remembered, moreover, that many poems, not composed by Map, were fathered upon him. Cp. the volume edited by Thomas Wright for the Camden Society, Latin Poems commonly attributed to Walter Map (London, 1841).

in the prime of his life the power to carry through a work of sustained invention, such as is exhibited in even the shortest of the three romances named above. He was one of the chief men of his age and of Welsh origin, and it was doubtless on these accounts that some scribe or redactor, or possibly even the author of some part of the cycle, endeavored to win the prestige of his name for these pseudo-Celtic stories.

7

Similar is the conclusion of Hinton, op. cit. p. 142. J. Bardoux in his Paris thesis, De Walterio Mappio, pp. 159 ff. (1900) had argued that Map supplied the material for the romances that are ascribed to him, although he did not write them himself.

8

Ward, p. 734 of the Catalogue cited above, quotes a passage from the Ipomedon, 11. 7183 (edition of E. Kölbing and E. Koschwitz, Breslau, 1889), of Hue de Rotelande to prove that Map really composed a romance, which he conjectures to have been a metrical Lancelot, of which our Vulgate Lancelot is the prose rendering. Hue, who wrote between 1174 and 1191, lived near Hereford, with which place Map was connected all his life, and he presumably knew Map. Shortly after a passage based on the three tournament motif in his Ipomedon, (the passage ends with 1. 6772) Hue excuses himself in a jesting manner for lying and says:

"Sul ne sai pas de mentir lart,

Walter map reset ben sa part."

Now this same three tournament motif is found in the prose Lancelot, III, 214ff., and Ward accordingly concludes that Hue, in writing his own passage on this motif, had in mind Map's treatment of the same subject in a hypothetical verse original of the prose Lancelot. Ward's theory has been taken up enthusiastically by Miss Weston, Three Days Tournament, pp. 6 ff. (London, 1902), and in her article on Map in the Encyclopaedia Britannica (Eleventh edition) also, by Brugger, Zs. f. frz. Spr. u. Litt. XXIX, 90 ff., (1905), both of whom identify this hypothetical verse Lancelot with the French original of Ulrich von Zatzikhoven's Lanzelet.

Hue's allusion to Map, as we have seen, occurs in a jesting passage, and the most natural interpretation of it would be that he was simply getting off a pleasantry at his friend's expense. Kölbing, p. VI of his edition, suggests that the allusion may be to Map's De Nugis. The three days tournament motif is so common a one in folk-tales and romances that the instances in the Ipomedon and Lanzelet need not be directly connected with each other. In any event, we certainly require something more solid than the proximity of this

motif and the allusion to Map in the Ipomedon (they are really separated by about 460 lines) to justify the hypothesis that Map wrote the lost French poem which is used at the beginning of the prose Lancelot. Generally speaking, the latter owes precious little to this lost French original of the Lanzelet (cp. RR, X, 54f., note), so that, even if we accepted Brugger's and Miss Weston's hypothesis, Map's contribution to the evolution of the cycle would be extremely slight.

H. Suchier, Zs. f. rom. Ph., XVI, 273 (1892) accepts the genuineness of the ascription of the cycle to Map on the ground that Manessier (Potvin VI, 158), who, in his opinion, wrote between 1214 and 1220 so shortly after Map's death had already before him the Queste with the same conclusion that we find in our extant MSS. about the original narrative of the Grail quest being preserved at Salisbury. As a matter of fact, Manessier here does not mention Map's name, though it stood very probably in his MS. of the Queste. If it did, this would merely prove that the ascription got into the MSS. very early, not that it was genuine. But the whole passage in question at the end of the Queste, including the ascription to Map, is, on the face of it, a fraud, for although the fact has been generally overlooked by Arthurian scholars, there was never any monastery at Salisbury. Cp. Dugdale's Monasticon Anglicanum and Abbot Gasquet's English Monastic Life, p. 301 (London, 1904), the list of English religious houses.

Suchier's views were adopted by E. Wechssler, Die Sage vom heiligen Gral, pp. 126ff. (Halle, 1898).

Birch-Hirschfeld, Sage vom Gral, pp. 234, ff., it should be observed, has called attention to the improbability of Map's always speaking of himself in the third person, as he is made to do in these ascriptions which we find in the MSS. of the Vulgate cycle. He suggests that such ascriptions got into the MSS. of that cycle through the confusion of some Meistre Gautier or other with the Walter, archdeacon of Oxford, who, according to Geoffrey of Monmouth, at the end of his Historia, brought the (fictitious) original of that work "ex Britannia". But this is not likely.

Of as little value as the passage in Manessier are the oft-quoted passages (cp. for instance, Birch-Hirschfeld, pp. 230f.) from the pretended Helie de Borron, viz. in the Prologue to Guiron le Courtois and in the Epilogue to the Bret, which speak of Map, "qui fu clers au roy Henri" (Prologue), as the author of "lestoire" (Prologue) or "lou propre livre" (Epilogue) "de monseigneur Lancelot dou Lac." There is no ground, however, for believing that this self-styled Helie de Borron, who certainly was not writing earlier than 1230, knew any more about the matter than we do. He was simply repeating or

drawing inferences from the ascriptions in MSS. like ours particularly, MSS. of the Queste and Mort Artu. This explains sufficiently what C. L. Kingsford in the article on Map in the Dictionary of National Biography regards as a confirmation of the correctness of Helie's statement, viz., the fact that he does not speak of Map as archdeacon the rank which Map held from the year 1197 on but as a simple clerk, which he was in his earlier life i. e. at the time when he was most likely to write romances.

The MS. of the Estoire printed by Hucher even assigns, III, 504, to Map a share in that romance. Robert wrote it with the "aid" of Map!